The 41-residue neuropeptide CRF originally characterized on the basis of its ability to activate the stress axis by triggering the release of cortisol in humans and corticosterone in rodents, has been demonstrated to mediate many effects of stress. CRF exhibits its actions through two G protein-dependent receptor subtypes, CRF1 and CRF2, which are differentially involved in the modulation of fear and anxiety formation. Learning measured by fear conditioning as conditioned fear is enhanced by activation of hippocampal CRF1 and impaired by activation of CRF2 of the lateral intermediate septum. Anxiety-like behavior is increased by activation of CRF1 - accessible through the brain ventricle system - or septal CRF2 and is reduced by activation of CRF2 or blockade of CRF1 -both accessible through the brain ventricles. Both CRF1 and CRF2 are involved in the response to a stressful stimulus. We observed in preliminary experiments that mice exposed to immobilization serving as stressful stimulus and subsequently, after increasing pauses, trained for fear conditioning, suffered from an intitial memory deficit from which they recovered within one hour after :he end of the stressful exposure. Interestingly, activation of a subpopulation of CRF2 lowered the anxiety-like behavior and prevented the memory deficit. We hypothesize that anxiety-like behavior and memory are inversely related. To obtain more insight into these behaviors, we now want to extend our analysis to defensive behaviors and locate the various receptor subpopulations with the help of ex vivo experiments. We will apply fear conditioning, the Elevated Plus Maze test, the Mouse Defense Test Battery and the Rat Exposure Test as behavioral paradigms to wild type mice and CRF1-deficient mice constitutively or conditionally lacking the CRF1 gene. We expect that this study will provide evidence that in the mouse model stress is linked to defined anxiety forms and that only selected stress-induced anxiety forms modulate the memory formation process as indicated by the fear conditioning results. It may well be that the cell biological and anatomical details of generating anxiety are more important for the impact on memory formation than the severity of the anxiety symptoms. On the basis of these considerations, this proposal is significant for psychiatric disorders such as PTSD and psychotic disorders in which stress modulates cognitive processes.